
import seaborn as sns # Seaborn is a data visualization library based on matplotlib, provides beautiful default styles and color palettes
import matplotlib.pyplot as plt # Matplotlib is a data visualization library in Python, used to create interactive plots and charts
import pandas as pd # Pandas is a library for data manipulation and analysis
import numpy as np # NumPy is a library for the Python programming language, adding support for large, multi-dimensional arrays and matrices


#Reading the movie data from a CSV file into a pandas dataframe
df = pd.read_csv('movies_initial.csv')

#Filtering out movies with less than 1000 votes on IMDB
df = df[df["imdbVotes"] >= 1000]

#Displaying the head of the filtered dataframe
df.head()


df.describe() #This function provides a statistical summary of the numerical columns in the dataframe, including count, mean, standard deviation, minimum and maximum values, as well as percentiles


df.dtypes #This function returns the data type of each column in the dataframe, which is useful for checking whether any columns need to be converted to a different data type before further analysis or visualization.

imdbID           int64
title           object
year            object
rating          object
runtime         object
genre           object
released        object
director        object
writer          object
cast            object
metacritic     float64
imdbRating     float64
imdbVotes      float64
poster          object
plot            object
fullplot        object
language        object
country         object
awards          object
lastupdated     object
type            object
dtype: object


#PrePro
import re
df['year'] = df['year'].apply(lambda x: re.sub('[^0-9]', '', x))
df["year"] = df["year"].str[:4]
df['year'] = df['year'].astype(int)
df["year"].value_counts()

year
2009    931
2013    895
2011    890
2008    888
2012    886
       ... 
1901      1
1888      1
1893      1
1913      1
1904      1
Name: count, Length: 121, dtype: int64


#PrePro
df = df.where(df["year"]>=1945)
df = df.where(df["year"]<=2016)


# In this code cell, we use Seaborn and Matplotlib libraries to visualize the number of missing values in each column of the filtered movie dataset.

import seaborn as sns  
import matplotlib.pyplot as plt 
# Counting the number of missing values in each column of the dataframe

missing_vals = df.isnull().sum()

# Creating a bar plot to show the number of missing values in each column

fig, ax = plt.subplots(figsize=(10, 5))
ax = sns.barplot(x=missing_vals.index, y=missing_vals.values)

# Adding labels and title to the plot

plt.xlabel('Columns')
plt.ylabel('Number of Missing Values')
plt.title('Number of Missing Values in each Column')

# Rotating the x-axis labels to improve readability

for item in ax.get_xticklabels():
    item.set_rotation(70)

# Displaying the plot

plt.show()


# Counting the number of unique values in each column of the dataframe

unique_vals = df.nunique()

# Creating a horizontal bar plot to show the number of unique values in each column

fig, ax = plt.subplots(figsize=(10, 5))
ax = sns.barplot(x=unique_vals.values, y=unique_vals.index, orient='horizontal')

# Setting the x-axis scale to logarithmic scale for improved visualization

ax.set_xscale('log')

# Adding labels and title to the plot

plt.ylabel('Columns')
plt.xlabel('Number of Unique Values (log scale)')
plt.title('Number of Unique Values in each Column')

# Displaying the plot

plt.show()


# Selecting the rows where 'imdbRating' column is not null
df_filt = df[df['imdbRating'].notnull()]

# Creating a figure with a larger width
fig, ax = plt.subplots(figsize=(10, 5))

# Plotting the distribution of the 'imdb_rating' column using Seaborn's histplot function
sns.histplot(data=df_filt, x="imdbRating", kde=True, ax=ax)

# Adding labels and title to the plot
plt.xlabel('IMDB Rating')
plt.ylabel('Density')
plt.title('Distribution of IMDB Ratings')

# Displaying the plot
plt.show()


# Counting the number of movies in each IMDB rating category and sorting them in ascending order
value_counts = df['imdbRating'].value_counts().sort_index()

# Printing the value counts of each IMDB rating category
value_counts

imdbRating
1.2    1
1.5    3
1.6    7
1.7    6
1.8    6
      ..
9.2    9
9.3    3
9.4    2
9.5    2
9.6    1
Name: count, Length: 83, dtype: int64


# Counting the number of movies in each genre category and sorting them in descending order
genre_value_counts = df['genre'].value_counts()
genre_value_counts = genre_value_counts.sort_values(ascending=False)

# Displaying the top 10 genres with highest number of movies
genre_value_counts.head(10)

genre
Drama                     1717
Comedy                    1120
Comedy, Drama              859
Drama, Romance             741
Comedy, Drama, Romance     720
Comedy, Romance            563
Horror                     449
Action, Crime, Drama       413
Horror, Thriller           384
Documentary                342
Name: count, dtype: int64


# Counting the number of movies in each genre category and sorting them in descending order
value_counts = df['genre'].value_counts()

# Selecting the top 10 genres based on number of movies
top_genres = value_counts.head(10).index.tolist()

# Filtering the dataset to include only movies with top 10 genres
df_top_genres = df[df['genre'].isin(top_genres)]

# Creating a violin plot to show the distribution of IMDB ratings for each top genre
fig, ax = plt.subplots(figsize=(20, 10))
sns.violinplot(x="genre", y="imdbRating", data=df_top_genres, ax=ax, showmedians=True)

# Adding labels and title to the plot
plt.xlabel('Genre')
plt.ylabel('IMDB Rating')
plt.title('IMDB Ratings by Genre (Top 10 by occurence in Dataset)')

# Rotating the x-axis labels to prevent overlap
for item in ax.get_xticklabels():
    item.set_rotation(45)

plt.show()


# Filtering out the rows where imdbRating is not available and year is between 1945 and 2016
#PrePro
df_mean = df[df['imdbRating'].notna()]
df_mean = df_mean.where(df_mean["year"]>=1945)
df_mean = df_mean.where(df_mean["year"]<=2016)

# Grouping the dataframe by year and computing the mean of imdbRating for each year
df_mean = df_mean.groupby('year')['imdbRating'].mean().reset_index()

# Create a line plot to visualize the trend of mean imdbRating over time
fig, ax = plt.subplots(figsize=(20, 10))
sns.lineplot(x="year", y="imdbRating", data=df_mean, ax=ax)

# Add labels and set the x-axis tick marks and labels
plt.xlabel('Year')
plt.ylabel('IMDB Rating Mean')
plt.title('IMDB Ratings Mean by Year')
xticks = range(int(df_mean['year'].min()), int(df_mean['year'].max()) + 5, 4)
ax.set_xticks(xticks)
ax.set_xticklabels(xticks)

plt.show()


fig, ax = plt.subplots(figsize=(20, 10))

# Create the regression plot
sns.regplot(x="year", y="imdbRating", data=df_mean, ax=ax)

# Add labels
plt.xlabel('Year')
plt.ylabel('IMDB Rating Mean')
plt.title('IMDB Ratings Mean by Year')

# Set the x-axis ticks and tick labels
xticks = range(int(df_mean['year'].min()), int(df_mean['year'].max()) + 5, 4)
ax.set_xticks(xticks)
ax.set_xticklabels(xticks)

# Rotate the x-axis tick labels
for item in ax.get_xticklabels():
    item.set_rotation(90)

# Display the plot
plt.show()


# Create a DataFrame of the top 250 movies by IMDB rating
df_top_250 = df.nlargest(250, 'imdbRating')

# Convert the year column to integers
df_top_250['year'] = df_top_250['year'].astype(int)

# Create a list of years to plot
years = range(int(df['year'].min()), int(df['year'].max())+1)

# Create a DataFrame with year as the index and count of movies as the values
df_years = pd.DataFrame({'year': years})
df_years['count'] = 0
df_years = df_years.set_index('year')

# Count the number of movies in each year
for year, group in df_top_250.groupby(['year']):
    df_years.loc[year, 'count'] = len(group)

# Plot the bar chart
df_years['count'].plot(kind='bar', figsize=(15, 5))
plt.xlabel('Year')
plt.ylabel('Number of Movies')
plt.title('Top 250 Movies by imdbRating per Year')
plt.gca().xaxis.set_major_locator(plt.MultipleLocator(5))
plt.show()


import plotly.express as px

# Create a bar chart with 'year' on the x-axis and 'imdbRating' on the y-axis
# Use the 'color' parameter to color-code the bars by 'imdbRating'
# Use the 'hover_data' parameter to include 'imdbVotes' as additional information when hovering over the bars
# Use the 'hover_name' parameter to display the 'imdbVotes' value when hovering over the bars
# Set the chart title
fig = px.bar(df_top_250, x='year', y='imdbRating', color='imdbRating',
             hover_data=['imdbVotes'],
             hover_name='imdbVotes',
             title="Top 250 movies distributed by year")

# Use the 'update_traces' method to add 'title' as an additional text to display when hovering over the bars
fig.update_traces(
                  hovertext=df_top_250['title'])

# Use the 'update_layout' method to hide the y-axis tick labels
fig.update_layout(
    yaxis=dict(
        showticklabels=False
    )
)

# Show the chart
fig.show()


import plotly.express as px

# Select the top 250 movies by imdbRating
df_top_250 = df.nlargest(250, 'imdbRating')

# Convert year to integer
df_top_250['year'] = df_top_250['year'].astype(int)

# Filter out movies with less than 25000 votes
df_top_250 = df_top_250[df_top_250["imdbVotes"] >= 25000]

# Create a bar chart with year on x-axis, imdbRating on y-axis, and color-coded by imdbRating
fig = px.bar(df_top_250, x='year', y='imdbRating', color='imdbRating',
             hover_data=['imdbVotes'], hover_name='imdbVotes',
             title="Top movies distributed by year (Real popular movies by no of votes)")

# Add movie title to hovertext
fig.update_traces(hovertext=df_top_250['title'])

# Remove y-axis tick labels
fig.update_layout(yaxis=dict(showticklabels=False))

# Show the plot
fig.show()


import scipy

# Get the top 10 most common genres
genre_counts = df["genre"].value_counts().head(10)
genres = genre_counts.index.tolist()

# Create a grid of subplots for each genre
fig, ax = plt.subplots(5, 2, figsize=(20, 30), sharex=True, sharey=True)
ax = ax.flatten()

# Loop through each genre and plot its mean IMDb rating over time
for i, genre in enumerate(genres):
    # Filter the data to only include rows for the current genre
    df_genre = df[df["genre"] == genre]
    
    # Filter out rows with missing IMDb ratings and limit the year range
    df_mean = df_genre[df_genre['imdbRating'].notna()]
    df_mean = df_mean.where(df_mean["year"]>=1945)
    df_mean = df_mean.where(df_mean["year"]<=2016)
    
    # Calculate the mean IMDb rating for each year
    df_mean = df_mean.groupby('year')['imdbRating'].mean().reset_index()
    
    # Create a scatter plot of the mean ratings over time
    sns.scatterplot(x="year", y="imdbRating", data=df_mean, ax=ax[i])
    
    # Add a red line to represent the linear regression line
    slope, intercept, r_value, p_value, std_err = scipy.stats.linregress(df_mean["year"], df_mean["imdbRating"])
    x = df_mean["year"]
    y = slope * x + intercept
    ax[i].plot(x, y, color="red")

    # Set the subplot title and axis labels
    ax[i].set_title(genre)
    if i == 8 or i == 9:
        ax[i].set_xlabel("Year")
    if i == 0 or i == 2 or i == 4 or i == 6 or i == 8:
        ax[i].set_ylabel("IMDB Rating Mean")

# Adjust the layout and display the plot
plt.tight_layout()
plt.show()


# Create a figure and axis for the plot
fig, ax = plt.subplots(figsize=(20, 10))

# Create a scatter plot of imdbVotes vs imdbRating
sns.scatterplot(x="imdbVotes", y="imdbRating", data=df, ax=ax)

# Calculate the correlation coefficient between imdbVotes and imdbRating
correlation_coefficient = df['imdbVotes'].corr(df['imdbRating'])

# Create a regression plot of imdbVotes vs imdbRating
sns.regplot(x="imdbVotes", y="imdbRating", data=df, ax=ax)

# Add text to the plot indicating the correlation coefficient
text = "Correlation coefficient: {:.2f}".format(correlation_coefficient)
ax.text(0.05, 0.95, text, transform=ax.transAxes, verticalalignment='top')

# Set the x and y axis labels and plot title
plt.xlabel('IMDB Votes')
plt.ylabel('IMDB Rating')
plt.title('IMDB Votes vs IMDB Rating')

# Show the plot
plt.show()


# Creating a scatter plot using the 'px.scatter()' function
fig = px.scatter(x=df["imdbVotes"],
                 y=df["imdbRating"],
                 hover_data=[df["title"], df["imdbRating"], df["year"]])

# Updating the layout of the scatter plot using the 'update_layout()' function.

fig.update_layout(
    title="IMDB Votes vs IMDB Rating",
    xaxis_title="IMDB Votes",
    yaxis_title="IMDB Rating",
    font=dict(size=10),
    width=1000,
    height=800
)

# Displaying the scatter plot using the 'show()' function.
fig.show()


#PrePro
#Dropping rows with missing values in the "country" column
df = df.dropna(subset=['country'])
#Converting all country names to lowercase, removing leading/trailing whitespaces,
#and splitting countries listed in a single row into a list of individual countries
df['country'] = df['country'].str.lower().str.strip().str.split(',')
#Sorting the list of countries for each movie in alphabetical order
df['country'] = df['country'].apply(lambda x: sorted(x))
#Joining the list of individual countries back into a single string with commas separating them
df['country'] = df['country'].apply(lambda x: ','.join(x))


# Get the top 30 countries by value counts
df_country = df['country'].value_counts().head(30)

# Store the value counts in a separate dataframe
occurece_df = df_country

# Get a list of the top countries
top_countries = df_country.index.tolist()

# Filter the original dataframe to only include rows with countries in the top countries list
df_top_countries = df[df['country'].isin(top_countries)]

# Group by country and calculate the mean imdbRating for each country
df_country = df_top_countries.groupby('country')['imdbRating'].mean().reset_index()

# Join the occurence_df to add the value counts as a new column
df_country = df_country.join(occurece_df, on=df_country["country"] , how='left', lsuffix='_left', rsuffix='_right')
# Sort the resulting dataframe by value counts in descending order
df_country.sort_values(by="country",inplace=True,ascending=False)

# Set the figure size for the plot
plt.figure(figsize=(20, 10))

# Create a barplot of average imdbRating by country using seaborn
sns.barplot(x="imdbRating", y="country", data=df_country,palette=reversed(sns.cubehelix_palette(n_colors=len(df_country))))

# Set the x and y labels and title for the plot
plt.xlabel("Average IMDB Rating")
plt.ylabel("Country")
plt.title("Average IMDB Rating per Country (Top 30), ordered by no of samples")

# Show the plot
plt.show()


df["awards"].value_counts()

awards
1 nomination.                                                1666
1 win.                                                        881
2 nominations.                                                682
1 win & 1 nomination.                                         608
3 nominations.                                                381
                                                             ... 
Nominated for 2 Oscars. Another 7 wins & 5 nominations.         1
Won 7 Oscars. Another 63 wins & 77 nominations.                 1
Nominated for 1 Oscar. Another 10 wins & 22 nominations.        1
Nominated for 5 Oscars. Another 16 wins & 69 nominations.       1
Nominated for 2 Oscars. Another 8 wins & 16 nominations.        1
Name: count, Length: 2728, dtype: int64


# Import regular expression module
import re

# Define a function to extract the total number of nominations and wins from a string
def extract_noms_and_wins(string):
    # Check if the input is a string
    if not isinstance(string, str):
        return 0
    # Use regex to find all integers in the string
    integers = re.findall(r'\d+', string)
    # Convert the list of strings to a list of integers
    integers = [int(i) for i in integers]
    # Return the sum of all integers in the list
    return sum(integers)

# Apply the function to the 'awards' column and store the result in a new column 'nomsandwins'
df['nomsandwins']= df['awards'].apply(extract_noms_and_wins)


# Import urllib module
import urllib

# Remove rows with missing values in the 'poster' column
df_removed = df.dropna(subset=['poster'])

# Get the top 10 films based on the nomsandwins column
top_10 = df_removed.nlargest(11, 'nomsandwins')
top_10.drop_duplicates(inplace=True, subset=["title"])

# Create a list of the URLs for the top 10 films
poster_urls = top_10['poster'].tolist()

# Download the images and store them as arrays
images = []
for url in poster_urls:
    with urllib.request.urlopen(url) as url:
        image = plt.imread(url, format='jpeg')
        images.append(image)

# Create a 2x5 grid of subplots to display the images
fig, ax = plt.subplots(2, 5, figsize=(20, 10))
for i, image in enumerate(images):
    # Display the image in the subplot
    ax[i//5, i%5].imshow(image)
    # Turn off the axis for the subplot
    ax[i//5, i%5].axis('off')
   
    # Add text to display the sum of nominations and wins below each image
    ax[i//5, i%5].text(140, -5, "Sum: "+ str(top_10.iloc[i]['nomsandwins']), 
                      ha='center', fontsize=15, color='black', fontweight='bold')
    # Add text to display the title of the movie above each image
    ax[i//5, i%5].text(140, 490, top_10.iloc[i]['title'].split(":")[0].strip(), 
                      ha='center', fontsize=15, color='black', fontweight='bold')

# Adjust the spacing between subplots
plt.subplots_adjust(hspace=0.3)

# Add a title for the entire figure
plt.suptitle("Top Movies by number of Won and Nominated Prices:", fontweight='bold', fontsize=16, y=0.93)

Text(0.5, 0.93, 'Top Movies by number of Won and Nominated Prices:')


# Create a barplot of the nomsandwins column
fig, ax = plt.subplots(figsize=(10, 5))
ax.barh(y=top_10['title'], width=top_10['nomsandwins'])

# Set the x and y labels for the plot
ax.set_ylabel("Title")
ax.set_xlabel("Number of Awards won or nominated")

# Adjust the layout of the plot
plt.tight_layout()

# Show the plot
plt.show()


# Get the top 30 countries by value counts
df_country = df['country'].value_counts().head(30)

# Store the value counts in a separate dataframe
occurece_df = df_country

# Get a list of the top countries
top_countries = df_country.index.tolist()

# Filter the original dataframe to only include rows with countries in the top countries list
df_top_countries = df[df['country'].isin(top_countries)]

# Group by country and calculate the mean nomsandwins for each country
df_country = df_top_countries.groupby('country')['nomsandwins'].mean().reset_index()

# Join the occurece_df to add the value counts as a new column
df_country = df_country.join(occurece_df, on=df_country["country"] , how='left', lsuffix='_left', rsuffix='_right')

# Sort the resulting dataframe by value counts in descending order
df_country.sort_values(by="country",inplace=True,ascending=False)

# Set the figure size for the plot
plt.figure(figsize=(20, 10))

# Create a barplot of average nomsandwins by country using seaborn
sns.barplot(x="nomsandwins", y="country", data=df_country,palette=reversed(sns.cubehelix_palette(n_colors=len(df_country))))

# Set the x and y labels and title for the plot
plt.xlabel("Average Prices nominated or won")
plt.ylabel("Country")
plt.title("Average Prices nominated or won per Country (Top 30), ordered by no of samples")

# Show the plot
plt.show()


# Import necessary modules
from wordcloud import WordCloud, STOPWORDS

# Add additional stopwords to the set of stopwords
STOPWORDS.add("one")
STOPWORDS.add("life")
STOPWORDS.add("live")
STOPWORDS.add("will")

# Get the top 5000 films based on imdbRating
df_top_5000 = df.nlargest(5000, 'imdbRating')

# Remove rows with missing values in the 'fullplot' column
df_top_5000 = df_top_5000.dropna(subset=['fullplot'])
#PrePro
# Concatenate all fullplot strings into one long string
text = ' '.join(df_top_5000['fullplot'].tolist())

# Create a set of stopwords
stopwords = set(STOPWORDS)

# Generate a word cloud from the text
wordcloud = WordCloud(stopwords=stopwords, background_color='white',
                      width=800, height=800, max_words=300,
                      min_font_size=10).generate(text)

# Set the figure size and facecolor for the plot
plt.figure(figsize=(8, 8), facecolor=None)

# Display the word cloud in the plot
plt.imshow(wordcloud, interpolation="bilinear")

# Turn off the axis for the plot
plt.axis("off")

# Add a title for the plot
plt.title("Word Cloud of Top 5000 Films by imdb Score\n", fontsize=20)

# Adjust the layout of the plot
plt.tight_layout(pad=0)
plt.show()


from collections import Counter

# Remove rows with missing values in the 'fullplot', 'genre', and 'imdbRating' columns
df_filtered = df.dropna(subset=['fullplot', 'genre',"imdbRating"])

#Remove multi genre movies
df_filtered['genre'] = df_filtered['genre'].str.split(',', expand=True)[0]

# Strip leading and trailing whitespace from the 'genre' column
df_filtered['genre'] = df_filtered['genre'].str.strip()

# Select only the 'fullplot', 'genre', and 'imdbRating' columns and sort by imdbRating in descending order
df_filtered = df_filtered[['fullplot', 'genre',"imdbRating"]].sort_values('imdbRating', ascending=False)

# Get the value counts for each genre
value_counts = df_filtered['genre'].value_counts()

# Get a list of the top 10 genres
top_genres = value_counts.head(10).index.tolist()
avg_imdb_score = df_filtered.groupby('genre')['imdbRating'].mean()
# Create a 5x2 grid of subplots to display the word clouds
fig, ax = plt.subplots(5, 2, figsize=(20, 30))
ax = ax.ravel()
for i, genre in enumerate(top_genres):
    # Get the texts for the current genre
    genre_text = df_filtered[df_filtered['genre'].str.contains(genre)]['fullplot'].str.cat(sep=' ')

    # Create a word cloud for the current genre
    wordcloud = WordCloud(width=800, height=800, background_color='white').generate(genre_text)

    # Display the word cloud in the subplot
    ax[i].imshow(wordcloud, interpolation='bilinear')
    # Turn off the axis for the subplot
    ax[i].axis('off')
    # Add a title for the subplot
    ax[i].set_title(f"{genre} (n={value_counts[genre]}, avg IMDb score={avg_imdb_score[genre]:.2f})")

# Add a title for the entire figure
plt.suptitle("Word clouds of top 10 genres of movies by imdb Score", fontsize=20, y=1.05)

# Adjust the layout of the plot
plt.tight_layout()

# Show the plot
plt.show()

C:\Users\swt4fe\AppData\Local\Temp\ipykernel_25020\3592765647.py:7: SettingWithCopyWarning:


A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy

C:\Users\swt4fe\AppData\Local\Temp\ipykernel_25020\3592765647.py:10: SettingWithCopyWarning:


A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy


import nltk
from nltk import word_tokenize
from nltk.tag import pos_tag
#PrePro
# Download necessary resources from nltk
nltk.download('punkt')
nltk.download('averaged_perceptron_tagger')

# Define a function to extract nouns from a text
def get_nouns(text):
    # Tokenize the text into words
    words = word_tokenize(text)
    # Tag the words with their part of speech
    tagged_words = pos_tag(words)
    # Extract the nouns from the tagged words
    nouns = [word for word, pos in tagged_words if pos in ['NN', 'NNS', 'NNP', 'NNPS']]
    # Return the list of nouns
    return nouns

# Create a copy of the original dataframe
df_nouns = df 

# Remove rows with missing values in the 'fullplot' column
df_nouns = df_nouns.dropna(subset=['fullplot'])

# Apply the get_nouns function to the 'fullplot' column and store the result in a new column 'fullplot_nouns'
df_nouns['fullplot_nouns'] = df_nouns['fullplot'].apply(get_nouns)

[nltk_data] Downloading package punkt to
[nltk_data]     C:\Users\swt4fe\AppData\Roaming\nltk_data...
[nltk_data]   Package punkt is already up-to-date!
[nltk_data] Downloading package averaged_perceptron_tagger to
[nltk_data]     C:\Users\swt4fe\AppData\Roaming\nltk_data...
[nltk_data]   Package averaged_perceptron_tagger is already up-to-
[nltk_data]       date!
C:\Users\swt4fe\AppData\Local\Temp\ipykernel_25020\2041100774.py:27: SettingWithCopyWarning:


A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy


# Get the top 500 films based on imdbRating
df_nouns = df_nouns.nlargest(500, 'imdbRating')

# Concatenate all fullplot_nouns lists into one long string
text = ""
for ele in df_nouns["fullplot_nouns"]:
     text = text + " ".join(ele)

# Generate a word cloud from the text
wordcloud = WordCloud(stopwords=stopwords, background_color='white',
                      width=800, height=800, max_words=300,
                      min_font_size=10).generate(text)

# Set the figure size and facecolor for the plot
plt.figure(figsize=(8, 8), facecolor=None)

# Display the word cloud in the plot
plt.imshow(wordcloud, interpolation="bilinear")

# Turn off the axis for the plot
plt.axis("off")

# Add a title for the plot
plt.title("Word Cloud of Top 500 Films by imdb Score (only nouns) \n", fontsize=20)

# Adjust the layout of the plot
plt.tight_layout(pad=0)

# Show the plot
plt.show()

	imdbID	title	year	rating	runtime	genre	released	director	writer	cast	...	imdbRating	imdbVotes	poster	plot	fullplot	language	country	awards	lastupdated	type
0	1	Carmencita	1894	NOT RATED	1 min	Documentary, Short	NaN	William K.L. Dickson	NaN	Carmencita	...	5.9	1032.0	https://m.media-amazon.com/images/M/MV5BMjAzND...	Performing on what looks like a small wooden s...	Performing on what looks like a small wooden s...	NaN	USA	NaN	2015-08-26 00:03:45.040000000	movie
1	5	Blacksmith Scene	1893	UNRATED	1 min	Short	1893-05-09	William K.L. Dickson	NaN	Charles Kayser, John Ott	...	6.2	1189.0	NaN	Three men hammer on an anvil and pass a bottle...	A stationary camera looks at a large anvil wit...	NaN	USA	1 win.	2015-08-26 00:03:50.133000000	movie
4	10	Employees Leaving the Lumi�re Factory	1895	NaN	1 min	Documentary, Short	1895-03-22	Louis Lumi�re	NaN	NaN	...	6.9	3469.0	NaN	A man opens the big gates to the Lumi�re facto...	A man opens the big gates to the Lumi�re facto...	NaN	France	NaN	2015-08-26 00:03:56.603000000	movie
5	12	The Arrival of a Train	1896	NaN	1 min	Documentary, Short	1896-01-01	Auguste Lumi�re, Louis Lumi�re	NaN	NaN	...	7.3	5043.0	https://m.media-amazon.com/images/M/MV5BMjEyND...	A group of people are standing in a straight l...	A group of people are standing in a straight l...	NaN	France	NaN	2015-08-15 00:02:53.443000000	movie
6	14	Tables Turned on the Gardener	1895	NaN	1 min	Comedy, Short	NaN	Louis Lumi�re	NaN	Fran�ois Clerc, Beno�t Duval	...	7.1	2554.0	NaN	A gardener is watering his flowers, when a mis...	A gardener is watering his flowers, when a mis...	NaN	France	NaN	2015-08-12 00:06:18.237000000	movie

	imdbID	metacritic	imdbRating	imdbVotes
count	2.214200e+04	7692.000000	22142.000000	2.214200e+04
mean	6.323814e+05	56.598284	6.496712	2.322801e+04
std	8.107426e+05	18.004372	1.182726	6.790026e+04
min	1.000000e+00	1.000000	1.200000	1.000000e+03
25%	8.884775e+04	44.000000	5.800000	1.805000e+03
50%	2.391470e+05	57.000000	6.700000	3.831000e+03
75%	1.029130e+06	70.000000	7.300000	1.311800e+04
max	4.727512e+06	100.000000	9.600000	1.521105e+06

Untersuchung eines IMDB Datensatzes¶

Zusammenfassung¶